Acute myocardial infarction (AMI) is a leading cause of death and disability globally [1]. In sub-Saharan Africa (SSA), the burden of AMI is rising rapidly as the region progresses through its epidemiological transition [2]. Approximately 250,000 deaths, or 5% of all deaths in SSA each year, are attributed to AMI [3]. In an effort to curb the rising global burden of AMI, the World Health Organization has prioritized evidence-based AMI treatment with antiplatelet agents as a “Best Buy” for reducing noncommunicable disease morbidity and mortality [4]. Yet, AMI care remains under-studied in SSA [2,5-7], and there are currently no published studies from the region evaluating quality improvement interventions to increase aspirin administration in AMI [8,9]. In Tanzania, recent studies have shown that uptake of evidence-based AMI care is suboptimal. Specifically, in a northern Tanzanian emergency department (ED), we found that only half of adult ED patients with potential AMI symptoms undergo testing with electrocardiogram (ECG) or cardiac biomarker testing, only 23% of patients with AMI were treated with aspirin, and only 5% of patients with AMI were taking aspirin 30 days following their diagnosis [10-12]. Against this backdrop, AMI outcomes in Tanzania are currently poor: approximately 90% of AMI cases are misdiagnosed, and 30-day mortality among patients with AMI is 43% [12].

In response to these findings, an interdisciplinary design team consisting of providers from the study hospital and global experts in implementation science, emergency medicine, and cardiology, developed a multicomponent intervention to improve the uptake of evidence-based AMI care in Tanzania. Because there are currently no published interventions to improve AMI care in SSA [8], the design team chose to adopt a quality improvement intervention from Brazil (the ACS-BRIDGE intervention) [13,14]. The team chose this intervention due to its emphasis on both diagnosis and treatment, its multicomponent nature, its appropriateness for a resource-limited setting, its preliminary efficacy, and its congruence with local barriers to care [15]. Guided by formative qualitative data [15], the design team tailored the ACS-BRIDGE intervention to the local Tanzanian context using the ADAPT-ITT framework [16]. The result was the tailored intervention, Multicomponent Intervention to improve acute Myocardial Infarction Care (MIMIC). This pilot trial will assess the implementation outcomes of MIMIC as well as its impact on clinical care in a Tanzanian ED.

This study will assess the acceptability, feasibility, and preliminary efficacy of the MIMIC intervention in a single-arm pilot trial. The primary outcome will be the acceptability of the MIMIC intervention. Secondary implementation outcomes will include feasibility and fidelity; secondary efficacy outcomes will include (1) the proportion of participants with chest pain or shortness of breath who undergo diagnostic testing for AMI, (2) the proportion of AMI participants who receive aspirin, (3) 30-day mortality among AMI participants, and (4) the proportion of AMI participants adhering to aspirin at 30 days following study enrollment.

The process of developing the MIMIC intervention has been described in detail elsewhere [17]. Briefly, an interdisciplinary design team considered a wide range of published interventions to improve AMI care [8,13,18-25]. Two existing quality improvement interventions for AMI clinical care were considered most suitable for potential adaptation to the resource-limited Tanzanian health care context. In India, the ACS-QUIK trial aimed to improve AMI care, but did not demonstrate a significant reduction in mortality or cardiovascular events among patients, included strategies (such as electronic discharge order sets) that were not possible in Tanzania, and emphasized rapid percutaneous coronary intervention, which is not currently available in Tanzania [21]. For these reasons, the ACS-QUIK study was not selected for adaptation. The other intervention, ACS-BRIDGE in Brazil, successfully bolstered adherence to evidence-based AMI therapies, reduced in-hospital major cardiovascular events, and included low-cost strategies that were more feasible in Tanzania, so it was selected for adaptation [13]. However, the health care landscape in Tanzania is much different from Brazil, with fewer resources and less access to advanced cardiac care [26]. To address this, the MIMIC intervention was adapted to the resource-limited context of Tanzania over a 1-year period of discussion and refinement by an international team of experts in emergency medicine, cardiology, and implementation science [17].

The MIMIC intervention is delineated in Figure 1. Briefly, the intervention consists of five key components: (1) a triage card that will be placed on the stretchers of patients with AMI symptoms by the triage nurse to prompt the physician to consider the diagnosis of AMI, (2) a pocket card summarizing the basic approach to AMI diagnosis and care that will be distributed to physicians and nurses, (3) a detailed web-based refresher module covering AMI diagnosis and care that will be required for all staff physicians and nurses, (4) educational pamphlets that will be distributed to patients with AMI and educational messages regarding AMI self-care that will be displayed on screens in the ED waiting room, and (5) a designated physician and nurse champion who will encourage the team and manage the delivery of all 5 intervention components. The full MIMIC intervention has been previously published. As described elsewhere, the patient education materials were piloted with patients from a range of educational backgrounds to ensure comprehensibility during the intervention development process [17].

The study will be conducted at Kilimanjaro Christian Medical Centre (KCMC), a tertiary care center in Moshi, Tanzania. KCMC is a zonal hospital covering a catchment area of approximately 15 million persons and was the site of our preliminary studies identifying gaps in evidence-based AMI care [10-12,15,27]. KCMC is currently equipped with basic AMI diagnostic tools, including ECG machines, both point-of-care and laboratory-based troponin assays, and echocardiograms. KCMC is also currently well-stocked with basic AMI medications, including aspirin, clopidogrel, heparin, statins, nitrates, beta-blockers, other antihypertensives, and thrombolytics. Additionally, KCMC offers a dedicated outpatient follow-up clinic for patients with cardiac conditions like ischemic heart disease. However, KCMC presently does not have a cardiologist on staff and lacks the capacity for percutaneous coronary intervention or coronary surgery.

This pilot trial will be conducted in the KCMC ED for 1 year. Implementation outcomes (acceptability, feasibility, and sustainability) will be assessed primarily via in-depth interviews and surveys conducted with key stakeholders (physicians, nurses, patients, and administrators). Secondary efficacy outcomes will be assessed via observation of the care of patients who present to the KCMC ED with possible AMI symptoms during the study period. Based on the results of prior surveillance studies in the KCMC ED [11,12,28], we expect to enroll approximately 400 patients with chest pain or dyspnea in the KCMC ED over 1 year.

For in-depth interviews and surveys assessing implementation outcomes, eligible participants will be KCMC staff, including hospital staff and physicians and nurses working in the ED during the study period, as well as patients diagnosed with AMI in the KCMC ED during the study period. Patients younger than 18 years and those who were unable to provide informed consent were excluded from study participation. For ED patients whose care will be observed to assess secondary efficacy outcomes, eligibility criteria will be as follows: (1) KCMC ED patient presenting during the study period, (2) being 18 years or older, and (3) primary or secondary complaint of chest pain or dyspnea. Exclusion criteria will be as follows: (1) chest pain secondary to trauma, (2) self-reported fever, and (3) inability to provide informed consent. Consistent with prior AMI studies in Tanzania [11,12,28], these exclusion criteria were selected to exclude patients whose chest pain or dyspnea is unlikely to be due to AMI.

For surveys administered to hospital staff to assess implementation outcomes, members of the research team will approach physicians, nurses, and administrators to offer them participation in the study. To minimize disruptions to clinical care, research staff will approach staff during break periods, meals, or at the conclusion of their shift. Staff will be compensated 5000 Tanzanian shillings (approximately US $2) for their time completing the survey.

For in-depth interviews evaluating implementation outcomes, purposive sampling will be used to ensure participants from diverse backgrounds are included. Participants will be hospital staff, administrators, and patients from a variety of occupations, ages, genders, and educational backgrounds. Selected hospital staff and administrators will again be approached by research staff during their break periods and offered participation in the in-depth interviews. Moreover, selected patients will be contacted via telephone and offered participation. All in-depth interview participants will be compensated for their time (approximately US $2).

The main objective of this trial is to assess the acceptability of MIMIC, which will determined by calculating the mean AIM score. AIM scores will be gathered from all participants, including patients, ED providers, and administrators, and then divided by the total number of trial participants. Importantly, the AIM scores from participants will be assumed to be normally distributed and independent of each other. The overall AIM score will fall within a range of 1 to 5, with higher scores indicating greater acceptability. An overall mean AIM score of 4.0 will be considered to indicate acceptability, as described above.

The acceptability of MIMIC will also be examined via qualitative analysis of the content of in-depth interviews across stakeholders. Two independent coders will conduct thematic analyses of interview transcripts using NVivo (version 14; Lumivero) in an iterative cycle of coding, discussion, and consensus-building. The coders will be assumed to be consistent and reliable in their coding process, and the findings from the qualitative analysis will be assumed to be accurate representations of participants’ views. Coders will use an inductive analytic approach, and disagreements among coders will be resolved by discussion and consensus-building. In the event a coding disagreement cannot be resolved by discussion, a third member of the coding team will serve as the tiebreaker. These thematic analyses will be guided by the Theoretical Framework for Acceptability [30].

Other implementation outcomes, such as fidelity and penetration, will be characterized by the proportion of participants receiving or complying with an intervention out of the total number of participants, as described above. In calculating these proportions, it will be assumed that the observations are independent of one another. Feasibility will be assessed through the FIM score mean and SD. To assess feasibility and costs implementation outcomes, the total number of resources allocated will be determined in place of proportions, as described above.

Efficacy outcomes will be used to assess the preliminary effectiveness of MIMIC. Pre-post analyses will be used to compare performance metrics from before and after MIMIC rollout to estimate the effect of the intervention on the 4 secondary efficacy outcomes listed in Table 2. Performance metrics from the preintervention and postintervention periods will be assumed to be independent of each other, and the conditions under which data were collected from each period will be assumed to be comparable.

To estimate effect sizes of the intervention, proportions of patients achieving each of the 4 key target efficacy outcomes will be compared from the preintervention period and postintervention period via Pearson chi-square test [48]. Preliminary effectiveness will be reported as odds ratios with corresponding 95% CIs constructed directly from two-by-two contingency tables. Preintervention data will be taken from an observational study conducted immediately prior to MIMIC rollout (February 1, 2023 to September 1, 2023). No a priori assumptions will be made regarding effect size of MIMIC on any of the 4 outcomes.

No a priori assumptions will be made about the acceptability of the MIMIC intervention. There are currently approximately 70 total physicians and nurses working in the KCMC ED, and we will aim to administer the provider survey to at least 90% of the staff (63 total participants). Recent studies of the acceptability of other health care quality improvement interventions using the AIM instrument have reported SDs ranging from 0.4 to 0.8 when reporting the AIM score on a 5-point scale [49-51]. Assuming an SD of 0.6 in our study, a sample size of 63 participants would allow us to estimate the mean AIM score with a margin of error of ±0.2 with 95% CI.

Our observational study of ED care during the MIMIC pilot trial will be conducted for 1 year. Based on prior observational studies in the KCMD ED [11,12], we expect to enroll 400 patients with chest pain or dyspnea during the trial. The data generated from this pilot study will inform sample size calculations for a future clinical trial.

For qualitative interviews, approximately 35 participants will be recruited. This cohort will include 10 patients with AMI, 10 physicians, 10 nurses, and 5 hospital administrators. Final sample size will be determined by thematic saturation.

Data from electronic health records and clinical care observations will be systematically collected on encrypted tablets using the Open Data Kit platform (version 2024.1.3; ODK) and uploaded directly to a secure, encrypted server at KCMC. Interview transcripts will be recorded and stored on encrypted digital devices accessible exclusively to authorized research personnel. Thematic analysis of these transcripts will be conducted using the NVivo software, with measures in place to encrypt and safeguard data. All data will be password-protected to maintain strict confidentiality. Prior to accessing the data within these systems, research personnel will undergo prerequisite training to ensure compliance with data security protocols. The Principal Investigator (JTH) will oversee data management by conducting weekly queries and reviewing inputted data for accuracy and completeness.

Following MIMIC rollout, stringent safety monitoring measures will be put in place to ensure the safety of the intervention. Importantly, MIMIC will be implemented by ED providers rather than by research staff, so the primary focus on safety monitoring will be to ensure that research activities do not disrupt the regular operations of the ED. Therefore, research personnel will undergo mandatory training to strictly observe and refrain from interfering with the clinical care provided to patients with AMI within the KCMC ED. Furthermore, all interviews will be scheduled during break periods, mealtimes, or at the conclusion of shifts to minimize any potential disruptions to patient care.

The principal investigator (JTH) and local coinvestigator (FMS) will be responsible for conducting safety reviews of the MIMIC intervention and addressing any threats to patient safety that may arise. Any instances of serious adverse events occurring during the implementation of MIMIC will be documented and reported to the local institutional review board in accordance with established safety and ethics protocols.

A formal interim analysis will be conducted 6 months into the pilot trial. At this point, the investigator team (JTH, GSB, NMT, FMS, JPB, and BTM) will analyze intervention fidelity data and patterns of care for four key care processes: (1) the proportion of patients presenting with chest pain or shortness of breath to the ED who undergo testing for AMI, (2) the proportion of patients with AMI who receive aspirin in the ED, (3) the proportion of patients with AMI who are alive at 30 days, and (4) the proportion of patients with AMI who are taking aspirin at 30 days. As this study involves a quality improvement intervention, we do not anticipate any adverse events. If any adverse events do occur, they will be reported, as required, to the Institutional Review Boards at Duke Health, KCMC, Tanzania National Institute for Medical Research, and the National Institutes of Health. We do not anticipate any changes to the trial protocol, but if these occur, these changes will be reported to the Institutional Review Boards at Duke Health, KCMC, Tanzania National Institute for Medical Research; the ClinicalTrials.gov record will also be updated.

Ethical approval for this study was obtained from the Tanzania National Institute for Medical Research (NIMR/HQ/R.8a/Vol. IX/2436, Version 7, February 23, 2021), KCMC (Proposal 893, Version 7, December 21, 2020), and Duke Health (Pro00090902, Version 1.7, September 16, 2020). All study procedures will be in accordance with the Helsinki Declaration of 1975, as revised in 2000. Research assistants will obtain written, informed consent from all research participants, including patients and ED providers, prior to their enrollment. Consent forms will be available in both English and Swahili to ensure they are understood by participants. Participation will be entirely voluntary, and patients all patients will reserve the right to opt out of the study at any time. Participant privacy will be respected, and no identifying information about participants will be shared publicly. Research personnel will observe MIMIC interventions and will refrain from disrupting clinical care. ED staff who participate in in-depth interviews and staff surveys will be compensated 5000 Tanzanian shillings (approximately US $2) in exchange for their time completing interviews and surveys. All data will be anonymized and stored in a secure, encrypted system accessible only to authorized research personnel. All clinically actionable findings identified via patient health records and observations of patient care will be reported to the primary clinical team at KCMC. In addition, all serious adverse events occurring during the implementation of MIMIC will be reported to the local institutional review board.

Findings of this pilot trial will be disseminated among KCMC ED providers and hospital administrators via in-person conference meetings and presentations. Findings will also be published in peer-reviewed journals to inform broader audiences of implementation science interventions to improve AMI care in resource-limited settings. Trial results will be shared with local and national representatives of the Ministry of Health in Tanzania. If MIMIC is found to be acceptable, the findings from this study will inform a stepped-wedge, cluster-randomized trial to assess the effectiveness and scalability of the intervention across multiple health care facilities in Tanzania.

In light of recent evidence suggesting that uptake of evidence-based AMI care is lacking in Tanzania, MIMIC holds promise as a valuable intervention that can improve AMI care delivery in the country. If MIMIC demonstrates successful implementation and efficacy outcomes in this pilot trial, then the results could pave the way for broader implementation of MIMIC in Tanzania and elsewhere in SSA, helping to mitigate the growing burden of AMI in the region.